Final Report : Investigation Group : CD - 01 PIs

We have completed a multi-year study of ecosystem carbon fluxes across Amazônia using the Simple Biosphere model (SiB2) at several spatial and temporal scales. We have developed understanding and confidence in the model simulations at local scales by direct evaluation against data collected in the field, and extrapolated to regional, Basin, and even global scales by coupling our local ecophysiological model to a range of atmospheric models. We have improved the model and developed input data from remotely sensed imagery at several scales. We have learned that water stress plays a less significant role in variations of ecosystem carbon flux than previously thought, and that interannual variations in the balance of C3 and C4 photosynthesis in the region contribute to substantial variability in the stable isotope ratio of terrestrial CO2 flux at the global scale. We have demonstrated that a handful of regular measurements of CO2 in the atmosphere above the region would dramatically improve the confidence of regional carbon budget estimates from global inversions. We have investigated the role of surface water and seasonally inundated land surfaces in the energy and carbon budgets of the region: preliminary evidence suggests that they play an important and previously neglected part in the Basin-scale carbon balance. Simulations of ecosystem responses to seasonal and interannual drought have demonstrated an important weakness of SiB: the model captures wet-season/dry-season carbon dynamics well in the Equatorial Tapajos region, but exaggerates the effects of drought stress under the more prolonged seasonal drought conditions in Rondonia. Exaggerated drought stress can produce catastrophic carbon and climate consequences in coupled models. Finally, we have tested a method for high-resolution regional CO2 flux estimation using an atmospheric mesoscale model.